ALMA MATER STUDIORUM UNIVERSITÀ DI BOLOGNA SCUOLA DI SCIENZE Corso di Laurea specialistica in Biologia Marina Outdoor production of Isochrysis galbana (T-iso) in industrial scale photobioreactors and modelling of its photosynthesis and respiration rate Tesi di laurea in Dinamica del fitoplancton Relatore: Presentata da: Prof. Rossella Pistocchi Davide Ippoliti Correlatore: Prof. Francisco Gabriel Acién Fernández II sessione Anno Accademico 2014/2015
73
Embed
ALMA MATER STUDIORUM UNIVERSITÀ DI BOLOGNAamslaurea.unibo.it/9591/1/Outdoor_production_of_Isochrysis_galbana... · Application of the photosynthesis rate model ... whereas the geometry
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
ALMA MATER STUDIORUM UNIVERSITÀ DI BOLOGNA
SCUOLA DI SCIENZE
Corso di Laurea specialistica in Biologia Marina
Outdoor production of Isochrysis galbana (T-iso) in
industrial scale photobioreactors and modelling of its
3.4. Measurement of photosynthesis and respiration rate .......... 23
3.5. Biomass concentration, fluorescence of chlorophylls and biochemical composition ............................................................. 26
3.6. Light utilization by the cultures .......................................... 27
3.7. Software and statistical analysis ......................................... 28
4. RESULTS AND DISCUSSIONS ............................................. 29
4.1. Photosynthesis and respiration rate models ........................ 29
4.2. Application of the photosynthesis rate model to industrial-scale T-PBRs ............................................................................... 44
4.3. Outdoor production of Isochrysis galbana (T-iso) ............. 49
Germany) and freeze-drying (Telstar Cryodos 50, Telstar, Spain)
culture samples at the steady state of each experiment. The protein
content was determined following a modification of the Lowry
method (López et al., 2010). The total lipid content was quantified
using the method proposed by Kochert (1978). Total ash was
determined by incineration of a 0.5 g sample in an oven at 450ºC
for 48 h. Finally, carbohydrates were estimated as the difference
(out of a hundred) after subtracting the lipid, protein and ash
content. To avoid interferences due to ash content the biochemical
composition is expressed on ash free basis.
27
3.6. Light utilization by the cultures
The biomass extinction coefficient (Ka), was calculated from the
average absorption value in the visible range (400-700 nm),
measured in a double beam Helios Alpha spectrophotometer. The
extinction coefficient was calculated by dividing the average
absorption by the biomass concentration (Cb) and the cuvette light
path (p) (Equation 1).
Ka =Abs!""!!""C! · p
Equation 1
The average irradiance at which cells are exposed inside a culture
(Iav) is a function of irradiance in the absence of cells (Io), the
biomass extinction coefficient (Ka), the biomass concentration (Cb)
and the light path inside the reactor (p). It can be approximated
using Equation 2 (Molina et al. 1997).
Iav =I!
K!C! · p· 1 − exp −K!C! · p Equation 2
Quantum yield (ΨE) is defined in microalgal cultures as the amount
of biomass generated by the unit of radiation (usually a mole of
photons) absorbed by the culture. Since it represents the ratio
between the biomass generated and the absorbed photon flux, it can
be calculated using Equation 3 (Molina et al. 1997) as a function of
the volumetric biomass productivity (Pb) and the photon flux
absorbed in the volume unit (Fvol). The photon flux absorbed
through the reactor volume can be obtained from the average
irradiance (Iav) on a culture volume; this can be calculated using
Equation 4 (Molina et al. 1997).
28
𝛹! =P!F!"#
Equation 3
F!"# = Iav · K! · C! Equation 4
Photosynthetic efficiency was calculated as the ratio of energy
stored in the biomass produced to energy impinging on the reactor
surface (Equation 5). The combustion heat of the biomass (H) was
calculated considering the specific caloric value of the lipids (38.9
kJ/g), proteins (24 kJ/g) and carbohydrates (16.6 kJ/g), and by
knowing the biochemical composition of the biomass. This
equation uses the volume to surface ratio (V/S) of the reactor, and
the PAR to global ratio of light, which was 2 E·MJ-1.
𝑃𝐸 =P! · H · VI · S
Equation 5
3.7. Software and statistical analysis
DaqFactory (Azeotech, USA) was used to control the
photobioreactors. Statistical analysis of the data was carried out
using the Statgraphics Centurion XVI software package. Non-linear
regression was used to fit experimental data to the proposed models
and to determine characteristic parameter values. Linear regression
was used to evaluate the influence of the dilution rate relatively to
the variables studied. Microsoft Excel was used to perform
simulations using the developed models.
29
4. RESULTS AND DISCUSSIONS
4.1. Photosynthesis and respiration rate models
Microalgae productivity is influenced by various factors but under
unlimited nutrient conditions, the most important are irradiance,
temperature, pH and dissolved oxygen. To optimize the
productivity of whichever strain in real-scale reactors, the most
accurate method is to expose it to different conditions directly in
industrial-scale photobioreactors. However, this would require high
costs and a long time. Alternatively, the influence of these factors
can be studied in the laboratory simulating outdoor conditions to
develop models that must subsequently be verified outdoors; most
of these models analyse the production of oxygen (photosynthesis
and respiration) as the first step in the biomass production process
(Costache et al., 2013; Yun and Park, 2003; Vejrazka et al., 2013),
this was the method used in the present work. Therefore, to model
the response of Isochrysis galbana cells to environmental factors,
experiments were performed measuring the oxygen production rate
under light conditions and the respiration rate under dark
conditions, with controlled environmental variables. These
measurements allowed us to calculate the cells’ net photosynthesis
from the difference between these two measurements. Samples
were collected from laboratory cultures operated in indoor
continuous mode but simulating outdoor conditions to maximize the
applicability of the developed models. Thus, experimental values of
biomass concentration and productivity measured indoors were
close to those measured outdoors, ranging from 0.6 to 0.9 gbiomass·l-1
30
and from 0.2 to 0.3 gbiomass·l-1·day-1. These values are higher than
those previously reported for this strain under outdoor conditions -
being 0.075 g·l-1·day-1 in tubular photobioreactors (van Bergeijk,
Salas-Leiton and Cañavate, 2010) and 0.13 g·l-1·day-1 in flat panels
(Zhang and Richmond, 2003), thus verifying the adequacy of the
culture conditions. Moreover, this strategy validates the
experiments themselves and their applicability in simulating real
outdoor cultures.
Concerning irradiance, it was observed that the net photosynthesis
rate increased from zero at zero irradiance up to values of 565
mgO2·gbiomass·h-1 at an irradiance of 600 µE·m-2·s-1, then decreased
due to photo-inhibition to 394 mgO2·gbiomass·h-1 at an irradiance of
2000 µE·m-2·s-1 (Figure 12A). According to these data, the
photosynthesis rate is maximal at an irradiance ranging from 500 to
1000 µE·m-2·s-1. Furthermore, it is confirmed that this strain is less
tolerant to high irradiance than other commercial strains such as
Haematococcus pluvialis, Chlorella vulgaris and Scenedesmus
almeriensis (Costache et al., 2013; Yun and Park, 2003; Jeon, Cho
and Yun, 2006). Nevertheless, the maximum daily irradiance to
which the T-PBRs were exposed during this research (April to
June) ranged from 600 to 900 µE·m-2·s-1, within the experimentally-
determined optimal value range. Concerning respiration, at zero
irradiance the respiration rate was 37 mgO2·gbiomass·h-1, increasing
with irradiance up to 111 mgO2·gbiomass·h-1 at an irradiance of 400
µE·m-2·s-1, then remaining constant whatever the irradiance in the
range tested (Figure 12B). This upward trend in the respiration rate
is also confirmed by other strains such as Coelastrum sphaericum
and Scenedesmus falcatus (Grobbelaar and Soeder, 1985). The
31
oxygen production rate is the difference between these two
parameters, representing the production of oxygen under light
conditions, which is proportional to the biomass production rate.
The data demonstrated that the oxygen production rate is negative
under dark conditions, with an irradiance level equal to 15 µE·m-
2·s-1 being necessary to achieve an oxygen production rate equal to
zero (the compensation point) (Figure 12C).
32
Fig. 12 - Influence of irradiance on the net photosynthesis (A), respiration (B) and oxygen production rate (C) of Isochrysis galbana at 29ºC, pH=7.5 and DO2=7.6 mg/L. Lines correspond to fit the proposed models (Equation 6, Equation 7).
33
At an irradiance higher than 15 µE·m-2·s-1, the net photosynthesis
rate is much higher than the respiration rate, thus the oxygen
production rate fits to the behaviour observed for the net
photosynthesis rate. According to these results, the maximal
respiration rate is 20% of the maximal photosynthesis rate; this ratio
being strain specific (Geider and Osborne, 1989). From these data it
could be concluded that the net photosynthesis rate can be fitted to
the Eiler and Peters model (Eq. 6) (Eilers and Peeters, 1988)
whereas the respiration rate can be fitted to the hyperbolic model
with no inhibition (Eq. 7).
c+b·I+a·II=PO2(I) 2
Equation 6
I+IkRO2max·I+RO2min=RO2(I) nn
n
Equation 7
The response of the net photosynthesis rate to irradiance is
modulated by other environmental conditions such as temperature,
pH and dissolved oxygen. The experiments carried out allow us to
calculate the normalized net photosynthesis rate as a function of
these culture conditions. Data showed that temperature and pH
exhibit similar behaviour, with the net photosynthesis rate being
zero at 12ºC and pH 3, increasing to a maximum of 36ºC and pH
7.5, then decreasing to zero at maximum values of 45ºC and pH 10
(Figure 13A,B). It was unexpected that the optimal temperature for
maximizing the photosynthesis rate was close to 36ºC. This
temperature is high when compared to previously reported optimal
temperatures for this strain, ranging from 25ºC to 30ºC (Marchetti
et al., 2012; Renaud et al., 2002; Claquin et al., 2008). Regarding
the pH, tolerance to this parameter is different for each strain. For
most microalgae growth it is optimal at pH values between 7.0 and
34
8.0; however for others such as Spirulina, alkaline pH is
recommended; pH has even been reported as useful in controlling
contaminants in mixed cultures (Goldman et al., 1982). Regarding
the dissolved oxygen concentration, the net photosynthesis rate is
maximal at dissolved oxygen concentrations from zero to 11 mg·l-1,
then exponentially decreases to zero at 20 mg·l-1 due to oxygen
inhibition (Figure 13C).
35
Fig.13 - Influence of temperature (A), pH (B) and dissolved oxygen concentration (C) on the normalized photosynthesis rate of Isochrysis galbana at 600 µE/m2·s. Experiments performed under standard conditions for other culture conditions (29ºC, pH=7.5 and DO2=7.6 mg/L). Lines correspond to fit the proposed models (Equation 8, Equation 9, Equation 10).
36
It’s important to note that this is a relevant parameter because
oxygen is produced during photosynthesis and can accumulate in
high concentrations in closed photobioreactors, reducing the
photosynthesis rate and favouring culture photorespiration
(Mendoza et al., 2013). Therefore, in high dissolved oxygen
concentrations, photorespiration takes place; the oxygen binds to
the Rubisco enzyme and modifies its role from carboxylase to
oxygenase, reducing carbohydrate synthesis (Badger et al., 2000).
Variation in the normalized net photosynthesis rate with
temperature and pH can be fitted to the cardinal model developed
for bacteria (Rosso et al., 1993) and validated for microalgae
(Bernard and Rémond, 2012). According to this model, the net
photosynthesis rate is a function of the difference between the
variables (temperature, pH) and the characteristic values of the
strain (maximal, minimal and optimal), defined only in the range of
tolerable values. Other authors proposed models based on the
Arrhenius equation (Costache et al., 2013; Pérez et al. 2008) or a
non-linear correlation (Blanchard et al., 1996; Moisan, Moisan and
Abbott, 2002) to consider the influence of temperature on the
microalgae cultures. Thus, equation 8 and equation 9 allow us to
model the response of the net photosynthesis rate to temperature
and pH according to the cardinal model (Bernard and Rémond,
2012). A model considering inhibition by product, as previously
reported (Costache et al., 2013) can be used to model the response
An analogous study was performed concerning the respiration rate.
The variation in the normalized respiration rate with the
temperature, pH and dissolved oxygen shows the same behaviour
(Figure 14), which was analogous to that previously observed for
the variation in the net photosynthesis rate with temperature and
pH.
38
Fig.14 - Influence of temperature (A), pH (B) and dissolved oxygen concentration (C) on the normalized respiration rate of Isochrysis galbana. Experiments performed under standard conditions for other culture conditions (29ºC, pH=7.5 and DO2=7.6 mg/L). Lines correspond to fit the proposed models (Equation 11, Equation 12, Equation 13).
39
The respiration rate was zero at values of 12ºC, pH 3 and 0 mg·l-1,
but it increased with the temperature, pH and dissolved oxygen to
be maximal at 32ºC, pH 7.5 and 16 mg·l-1, then decreased to zero at
maximal values of 46ºC, pH 10 and 26 mg·l-1. The influence of
temperature and pH on the respiration rate is related to the
adequacy of the culture conditions to the optimal ones required by
the strain, whereas the influence of dissolved oxygen concentration
is related to the nutrient availability or excess; in this case, the
dissolved oxygen required for respiration. According to these
results, the variation in the normalized respiration rate to changes in
temperature or pH agree with the previously-reported cardinal
model, whereas the influence of dissolved oxygen can be modelled
using nutrient limitation-inhibition models. However, because the
cardinal model also reproduces the observed pattern, and to reduce
the variety of the equations used, this cardinal model has also been
employed to model the influence of dissolved oxygen on the
respiration rate. Thus, the following equations are used to model the
respiration rate’s response to temperature (Eq. 11), pH (Eq. 12) and
The proposed equations allows us to model the photosynthesis and
respiration rates as a function of the culture conditions the cells are
exposed to inside the culture over a short period, but possibly
conditions that are not supported over long periods, or whose
performance modifies as a function of exposure time. To determine
the “validity time” of the model, it is necessary to evaluate the
tolerance of Isochrysis galbana cells to more adverse culture
conditions. From the experimental data, the most unexpected value
was the optimal temperature of 36ºC. It is possible that although the
photosynthesis rate increases with temperature over a short period,
over longer periods other adverse effects occur at such a
temperature and, thus, the overall performance of the cells
diminishes. To study the model validity time, a culture was
continuously exposed to a temperature of 36ºC, samples being
taken at different times to study the variations in the photosynthesis
and respiration rates over time. Data show that the net
photosynthesis rate was stable for at least 100 minutes, then slightly
decreased after 150 minutes (Figure 15A), thus confirming the
tolerance of this strain to high temperatures for a maximal two-hour
period. Regarding the respiration rate, it was constant the whole
time, there was no observable tendency for exposure time to high
temperature (Figure 15B). Because the net photosynthesis rate was
much higher than the respiration rate under the standard conditions
used, the oxygen production rate behaved similarly to that discussed
for the net photosynthesis rate (Figure 15C).
41
Fig.15 - Variation in the net photosynthesis rate (A), respiration rate (B) and oxygen production rate (C) with time for Isochrysis galbana cells exposed to high temperature (35ºC) for a long period. Measurements performed under the standard culture conditions (I=600 µE/m2·s , pH=7.5, DO2=10 mg/L).
0
100
200
300
400
500
600
700
800
900
0 50 100 150 200 250 300
Net pho
tosynthe
sis rate (m
gO2/gbiomass·∙h)
Time (min)
A)
0
20
40
60
80
100
120
140
160
0 50 100 150 200 250 300
Respira
tion rate (m
gO2/gbiomass·∙h)
Time (min)
B)
0
100
200
300
400
500
600
700
800
0 50 100 150 200 250 300
Oxygen prod
uctio
n rate (m
gO2/gbiomass·∙h)
Time (min)
C)
42
According to these results, the net photosynthesis and respiration
rates can be modelled by combining these equations to obtain a
general equation representing the overall strain behaviour based on
the observed patterns. Thus, equation 14 and equation 15 allow us
to modelled the net photosynthesis and respiration rates as a
function of the culture conditions (irradiance, temperature, pH and
dissolved oxygen) to which the cells are exposed.
PO2(DO2)·PO2(pH)·PO2(T)PO2(I)·=PO2 Equation 14
RO2(DO2)·RO2(pH)·RO2(T)RO2(I)·=RO2 Equation 15
In order to determine the optimal values of the characteristic
parameters included in these models, we performed a non-linear
regression of the entire data to the experimental data set.
Table 1.- Values for the proposed model’s parameter characteristics (Equations
6-15) obtained by non-linear regression of the experimental values of the net
photosynthesis rate and respiration rate under the experimental conditions
tested.
Net photosynthesis rate Respiration rate Parameter Value Units Parameter Value Units a 3.42E-07 RO2min 52.17 mgO2/L·h b 9.30E-04 RO2max 153.00 mgO2/L·h c 2.90E-01 Ik 1152.00 µE/m2·s PO2max 641.01 mgO2/L·h n 1.90 Ik 186.05 µE/m2·s Tmin 12.84 ºC Im 921.23 µE/m2·s Tmax 45.82 ºC alfa 3.45 Topt 33.00 ºC Tmin 11.88 ºC pHmin 3.00 Tmax 46.15 ºC pHmax 10.00 Topt 35.73 ºC pHopt 7.50 pHmin 2.24 DO2min 0.70 mgO2/L pHmax 10.00 DO2max 23.27 mgO2/L pHopt 7.34 DO2opt 11.96 mgO2/L KO2 19.99 mgO2/L z 2.90
43
The value of characteristic parameters obtained is shown in table 1
whereas the correlation between experimental and simulated values
obtained using the proposed model (Equation 6-15) and the
characteristic parameter values obtained is shown in figure 16.
Fig.16 - Correlation between experimental and simulated values of the net photosynthesis rate (A) and respiration rate (B) of Isochrysis galbana. Simulated values obtained using the proposed models (Equation 6-15) and the parameter characteristic values shown in table 1.
One can observe how the proposed model simulates the
experimental values of the net photosynthesis and respiration rates;
with the correlation for the net photosynthesis rate being even
y = 0.9777xR² = 0.896
0
100
200
300
400
500
600
700
0 100 200 300 400 500 600 700
Simulated
net pho
tosynthe
sis rate
(mgO
2/gbiomass·∙h)
Experimental net photosynthesis rate (mgO2/gbiomass·∙h)
A)
y = 0.9255xR² = 0.6116
0
20
40
60
80
100
120
140
0 20 40 60 80 100 120 140
Simulated
respira
tion rate
(mgO
2/gbiomass·∙h)
Experimental respiration rate (mgO2/gbiomass·∙h)
B)
44
higher than that for the respiration rate. According to these results,
it is demonstrated that the developed model reproduces the
photosynthesis and respiration rates of Isochrysis not only for short
time periods (minutes) but also for long periods (up to two hours);
thus making them sufficiently robust to be exposed to real outdoor
conditions where changes in the culture conditions, especially
temperature, take place slowly in line with the solar cycle.
4.2. Application of the photosynthesis rate model to
industrial-scale T-PBRs
The model has been validated to study the performance of
Isochrysis galbana cultures grown in an industrial-scale tubular
photobioreactor. For this, experimental data of the average
irradiance, temperature, pH and dissolved oxygen measured on-line
in the reactor during a solar cycle in steady-state has been used
(Figure 17A). Data show that the average irradiance inside the
culture ranges daily from zero to 1000 µE·m-2·s-1 confirming that
the cells are mainly light-limited, even at noon when there is high
solar irradiance (up to 1600 µE·m-2·s-1). Moreover, during the solar
cycle, the temperature ranges from 19.0ºC to 29.7ºC, whereas the
pH varies from 7.6 to 8.1, and the dissolved oxygen from 5.1 to
14.4 mg·l-1. These values are close to those previously reported for
outdoor cultures of Isochrysis in outdoor photobioreactors (van
Bergeijk, Salas-Leiton and Cañavate, 2010; Zhang and Richmond,
2003). Concerning temperature variation, this modifies by up to
10ºC, according to the solar cycle, because the control system only
allows us to avoid overheating of the culture inside the reactor.
45
Furthermore, the average value was 24.3ºC, far from the optimal
photosynthesis rate value determined at 36ºC (Figure 13A) but
close to the previously-reported optimal rate, at 21ºC (van Bergeijk,
Salas-Leiton and Cañavate, 2010). The temperature effect is very
important on outdoor microalgal cultures subject to daily culture
condition variations (particularly irradiance and temperature) for
which simultaneous adverse conditions can take place (van
Bergeijk, Salas-Leiton and Cañavate, 2010). The negative effect of
low temperature and high irradiance has already been demonstrated
when occurring in the first hours of the morning in outdoor raceway
ponds, with these conditions enhancing the photoinhibition
phenomena. The utilization of closed photobioreactors rather than
open raceways allows us to improve temperature control, making it
possible to increase the night-time temperature and to avoid
overheating at noon - but this involves a higher production cost
(Acién et al., 2013). The variation in pH is minimal, up to 0.5 pH
units, due to the control system’s adequacy in supplying CO2 and
controlling pH. The average pH value is 7.9, close to the optimal
value determined for photosynthesis rate of Isochrysis T-iso, of 7.5
(Figure 13B). Microalgae culture productivity can be influenced not
only by the mean pH value but also by local pH gradients that take
place when pure carbon dioxide is supplied (Fernández et al.,
2014). The data obtained show the optimal pH range is between 7.0
and 8.0 - therefore to optimize productivity in the T-PBRs, it was
enough to use the on-demand injection of pure CO2 to keep the pH
lower than 8.0. Finally, the average dissolved oxygen concentration
was 9.1 mg·l-1, close to air saturation, but at noon, values up to 14.4
mg·l-1 were measured due to the system’s inability to remove all the
46
oxygen produced by photosynthesis. According to the
mg·l-1 reduce cell performance (Figure 13C), and therefore
represent a relevant reduction in productivity.
From these experimental values of temperature, pH and dissolved
oxygen, the normalized photosynthesis rate can be calculated by
Equation 8 to Equation 10 (Figure 17B). Results confirm that pH
was adequately controlled and did not reduce the photosynthesis
rate, whereas temperature and dissolved oxygen varied greatly
through the solar cycle, values being far from optimal, and thus the
normalized photosynthesis rate was less than maximal. A larger
contribution to a reduced photosynthesis rate came from the
inadequacy of the culture conditions because of temperature and
dissolved oxygen deviations taking place during the light period.
The integral of values during the light period was 0.97 for
PO2(pH), whereas it was 0.61 for PO2(T) and 0.77 for PO2(DO2).
These results indicate that the photosynthesis rate can be increased
39% by optimizing the temperature control and 27% by optimizing
the dissolved oxygen concentration inside the culture - thus
photosynthesis can be more than doubled by optimizing both
parameters at the same time (PO2(T-pH-DO2) being 0.44).
To better understand these phenomena, figure 6C shows the
expected photosynthesis rate based on average irradiance inside the
culture (considering optimal values of temperature, pH, and
dissolved oxygen) (Eq. 6), and calculated considering all the culture
conditions (irradiance, temperature, pH and dissolved oxygen) (Eq.
14). It can be observed that net photosynthesis, as a function solely
of irradiance (PO2(I)), is much higher than that calculated when
47
considering all the culture parameters (PO2(I-T-pH-DO2)),
especially at noon when light availability is maximal. Thus, the
integral of PO2(I) values allows us to calculate the maximal oxygen
production, 0.92 gO2·l-1·day-1, whereas the integral of PO2(I-T-pH-
DO2) allows us to calculate the real oxygen production, 0.42 gO2·l-
1·day-1. In this way, it is confirmed that the photosynthesis rate can
be doubled by optimizing the control of the culture conditions in the
reactors used.
48
Fig.17 - Daily variation in culture parameters of Isochrysis galbana culture in an industrial-scale tubular photobioreactor operated in continuous mode at 0.4 1/day. A) Experimental values of average irradiance, temperature, pH and dissolved oxygen to which the cells are exposed inside the culture; B) Influence of culture conditions (T, pH, DO2) in the normalized photosynthesis rate according to the proposed equations (Equation 13-15); C) Net photosynthesis rate as a function of irradiance (Equation 6) and of all the variables (I, T, pH, DO2) (Equation 14).
0
5
10
15
20
25
30
0
10
20
30
40
50
60
70
80
90
100
0 4 8 12 16 20 24
Dissolved oxygen
(mgO
2/L), p
H
Irrad
iance (m
icroE/m2·∙s), Tem
perature (ºC)
Hour
Iav
T
pH
DO2
A)
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0 4 8 12 16 20 24
Normalized
net pho
tosynthe
sis rate
Hour
PO2(T) PO2(pH)
PO2(DO2) PO2(T·∙pH·∙DO2)
B)
0
20
40
60
80
100
120
140
0 4 8 12 16 20 24
Net pho
tosynthe
sis rate (m
gO2/L·∙h)
Hour
PO2(I)
PO2(I·∙T·∙pH·∙DO2)
C)
49
Moreover, by comparing the real oxygen production, calculated
using the proposed model, with the experimental biomass
productivity measured, of 0.22 gbiomass·l-1·day-1, the oxygen to
biomass ratio can be calculated a value of 1.94 gO2·gbiomass-1. In
outdoor cultures of Scenedesmus almeriensis carried out in open
raceway reactors, an oxygen-to-biomass ratio of 1.46 gO2 gbiomass-1
was measured when using flue gas to control pH and to supply CO2;
whereas a ratio of 0.99 gO2 gbiomass-1 was measured when using pure
CO2 (Mendoza et al., 2013). The oxygen yield value reported here
agrees with that previously reported, thus confirming the proposed
model’s validity in determining oxygen production as a function of
culture conditions; and, moreover, loss of productivity due to
deviation in the culture conditions. It is therefore a useful tool for
taking decisions regarding the implementation of different control
strategies.
4.3. Outdoor production of Isochrysis galbana (T-iso)
To determine the feasibility of producing Isochrysis galbana T-iso
in outdoor industrial scale tubular photobioreactors experiments
were performed by modifying the dilution rate under standard
conditions. Experiments were performed in April-May (2015), the
cultures performing adequately at the dilution rates tested.
Subsequently, to evaluate the influence of the various factors linear
correlations were made between the various parameters and the
dilution (Table 2).
50
Tab. 2 – P-Values obtained by linear correlations (Pearson) between the
parameters and the rate of dilution
Results show as increasing the dilution rate from 0.15 to 0.35 1/h
the biomass concentration on steady state diminishes from 1.1 to
0.7 g/L (P-Value < -0.05: significant difference, Tab. 2), but the
chlorophyll fluorescence remained constant, at values of 0.58 to
0.65 (P-Value > 0.05: no significant difference, Tab. 2) , in all the
experiments (Figure 18A). The reduction of biomass concentration
with the increase of dilution rate represents the behaviour expected
from light limited cultures, whereas the maintenance of
fluorescence of chlorophylls with the increase of dilution rate
indicates that this variable has not influence on the status of the
photosynthetic apparatus. Regarding biomass productivity, it
increased with the increase of dilution rate (P-Value < 0.05:
significant difference, Tab. 2), thus confirming that the cultures
were light limited under the imposed culture conditions, and that
higher biomass productivity could be reached by operating at higher
dilution rates (Figure 18B). However, the dilution rate was not
Parameter P-Value Parameter P-Value Biomass concentration --0.003528 pH minimum 0.238610 Fluorescence of chlorophylls 0.253081 Temperature average 0.547727 Volumetric productivity 0.006219 Temperature maximum -0.037913 Areal productivity 0.006219 Temperatura minimum 0.138590 Extinction coefficient 0.005737 Average irradiance 0.007631 Dissolved oxygen average 0.097486 Protein content 0.002218 Dissolved oxygen maximum 0.142337 Lipid content -0.070122 Dissolved oxygen minimum 0.042923 Carbohydrate content -0.082813 pH average 0.045145 Photosynthetic efficiency 0.015367 pH maximum 0.090452 Quatum yield 0.006739
51
increased over 0.35 1/day because the biomass concentration in
steady state was lowered below a critical value of 0.8 g/L. In this
case the T-iso cultures in outdoor conditions can be unstable. The
maximal biomass productivity achieved was 0.25 g/Lday equivalent
to 20 g/m2·day due to the V/S ratio of the photobioreactors utilized.
With respect to the light availability, results shows as increasing the
dilution rate the extinction coefficient of the biomass increased, at
the same time the average irradiance inside the culture also
increased (both P-Values < 0.05: significant difference, Tab. 2), in
spite of constant solar radiation (Figure 18C).
52
Fig.18 - Variation of (A) biomass concentration and fluorescence of chlorophylls, (B) volumetric and areal productivity, and (C) extinction coefficient and average irradiance inside the culture, with the dilution rate, of Isochrysis galbana T-iso cultures performed in 3.0 m3 industrial scale tubular photobioreactors.
53
Thus, on the days when the experiments were performed the solar
radiation was stable, daily average radiation ranging from 350 to
360 W/m2. These results confirmed that the cultures were not
photo-limited, and that photo-limitation was not present or was
negligible. Thus, the extinction coefficient increased from 0.23 to
0.30 m2/g, whereas the average irradiance increased from 17 to 22
µE/m2·s when the dilution rate was increased from 0.15 to 0.35
1/day.
To determine if the productivity of the cultures was negatively
affected by the culture conditions, the experimental dissolved
oxygen, pH and temperature values were registered and analysed.
Maximum, minimum and average daily values are shown in Figure
19.
54
Fig. 19 - Variation of (A) dissolved oxygen, (B) pH, and (C) temperature into the culture, with the dilution rate, of Isochrysis galbana T-iso cultures performed in 3.0 m3 industrial scale tubular photobioreactors.
0
100
200
300
400
0.10 0.20 0.30 0.40
Dissok
ved oxygen
(%Sat.)
Dilution rate (1/day)
AverageMaximumMinimum
A)
0.0
2.0
4.0
6.0
8.0
10.0
0.10 0.20 0.30 0.40
pH
Dilution rate (1/day)
AverageMaximumMinimum
B)
0
10
20
30
40
0.10 0.20 0.30 0.40
Tempe
rature (ºC)
Dilution rate (1/day)
AverageMaximumMinimum
C)
55
Data show as the dissolved oxygen largely modifies along the day,
ranging from minimum values of 70%Sat during the night to more
than 300%Sat at noon, the average daily value being 170%Sat.
However, no clear influence of imposed dilution rate was observed
(Tab. 2). According to these results the reactors mass transfer
capacity was not enough to remove all the oxygen produced at
noon, thus photo-respiration phenomena can occur and diminish the
productivity of the cultures. Regarding the pH, variations along the
day resulted really low, lower than 0.5, with no differences being
observed as a function of dilution rate (Tab. 2). Finally, regarding
temperature, data shows as the temperature modifies along the day
from 20ºC during the night to maximum of 31ºC at noon. No
heating systems were used but the cultures increased its temperature
with the solar radiation due to the absorption of heat by radiation.
This heat absorption could increase the temperature over 40ºC, but
the reactors had a cooling system to prevent overheating and death
of the cultures, and the system worked adequately according to the
results. These experimental values are close to those previously
reported for cultures of Isochrysis in outdoor photobioreactors (van
Bergeijk et al., 2010; Zhang et al., 2003).
The biochemical composition of the biomass was analysed. To
avoid the ash content influence the data are expressed in ash free
basis. Results show as the dilution rate modifies the biochemical
composition of the biomass: the higher the imposed dilution rate the
higher was the protein content (P-Value <0.05 : significant
difference, Tab.2). It was also found a correlation with the decrease
of the lipid content, although not statistically significant (P-Value =
-0.07 : no significant difference, Tab. 2) (Figure 20).
56
Fig. 20 - Variation of biochemical composition (protein, lipids and carbohydrates) of the biomass with the dilution rate, of Isochrysis galbana T-iso cultures performed in 3.0 m3 industrial scale tubular photobioreactors.
Protein content increased from 35 to 45% when the dilution rate
increased from 0.15 to 0.35 1/day, the lipids content reducing from
28 to 20%. Carbohydrate content did not show a clear tendency
with the dilution rate variation, it represents between 28 to 30% of
the biomass. The protein and lipid content variation with increasing
the dilution rate is an expected behaviour due to the faster growth
rate of the cells, thus requiring more proteins and storing less
energy as lipids. However, it is very important to be capable to
produce whatever strain at stable dilution rate, as this allows to
maintain a stable biochemical composition of the biomass.
Tubular photobioreactors are recommended as the most suitable for
the outdoor production of microalgae strains which can be easily
contaminated, such as Isochrysis galbana (T-iso). On these reactors
it is possible to provide a better control of culture conditions, thus
being reliable to produce microalgae in continuous mode. However,
57
the tolerance of whatever strain to grown in this type of reactors
must be tested under real conditions. Thus, although on these
reactors different mechanisms can be used to control the pH,
dissolved oxygen and temperature, the daily variation of these
parameters along the solar period and including along the reactor
due to the existence of gradients, can diminish the productivity or
simply kill the cells.
The performed experiments demonstrated that Isochrysis galbana
(T-iso) can be produced in industrial size tubular photobioreactors
in continuous mode. The results indicate that the cultures mainly
perform as light limited cultures, thus the biomass productivity
increase by increasing the dilution rate. The chlorophyll
fluorescence remained constant whatever the imposed dilution rate,
therefore it was not observed photosynthetic apparatus stress. The
biomass showed an adequate and stable composition, being rich in
proteins and lipids, the most valuable components when used as
feed in aquaculture.
The productivity measured during experiments, of 0.16-0.23
g/L·day, is in the range previously reported for this strain or even
higher. Thus, at indoor conditions biomass productivities from 0.2
to 0.3 gbiomass·l-1·day-1 has been reported (in this study) whereas at
outdoor conditions a value of 0.075 g·l-1·day-1 has been reported in
tubular photobioreactors (van Bergeijk et al. 2010), and 0.13 g·l-
1·day-1 in flat panels (Zhang,C.W. 2003). However, these
productivities are much lower than those reported in tubular
photobioreactors using different strains. This can be due to the
lower growth rate of Isochrysis galbana T-iso versus other strains
or to the higher influence of culture deviation conditions versus the
58
optimal one for this microalga. By using the proposed model and
considering the culture parameters variation (dissolved oxygen, pH
and temperature) inside the reactors along the day, the normalized
photosynthesis rate can be calculated. Results demonstrate that the
pH was adequately controlled in the reactor, the normalized
photosynthesis rate being 0.97 at whatever dilution rate (Figure 21).
Fig. 21 - Influence of deviation of culture conditions versus the optimal reported for Isochrysis galbana T-iso into the normalized photosynthesis rate at different dilution rates tested. Data from experiments performed in 3.0 m3 industrial scale tubular photobioreactors.
The optimal pH for this strain was reported to be 8.0, although the
tolerance to deviations of this parameter are different according to
the strain. For most microalgae the growth is optimal at pH values
between 7.0 and 8.0; however for others as Spirulina alkaline pH is
recommended, considering that the high pH has been reported to be
useful for the control of contaminants in mixed cultures (Goldman
1982). However, dissolved oxygen has a larger variation during the
day, reaching values higher than 150 %Sat. that has been reported
as maximal tolerable for this strain without reduction of
0.0
0.2
0.4
0.6
0.8
1.0
1.2
0.10 0.20 0.30 0.40
Normalized
pho
tosynthe
sis rate
Dilution rate (1/day)
Dissolved oxygenpHTemperatureOverall (DO2*pH*T)
59
photosynthesis capacity. On this way the normalized photosynthesis
rate due to dissolved oxygen concentration was in the range of 0.7-
0.8, thus indicating that inadequate dissolved oxygen concentration
reduces the performance of the cultures between 30 and 20%. It has
been reported that the net photosynthesis rate is maximum at
dissolved oxygen concentration from zero to 11 mg·l-1, then
exponentially decreasing to reach zero at 20 mg·l-1 due to inhibition
by oxygen. It’s worth noticing that this is a relevant parameter
because oxygen is produced during photosynthesis and it can
accumulate in high concentrations in closed photobioreactors,
reducing the photosynthesis rate and favouring the photorespiration
of the cultures (Mendoza et al., 2013). Therefore, in high dissolved
oxygen concentrations photorespiration takes place; the oxygen
binds to the Rubisco enzyme and modifies its role from carboxylase
to oxygenase, reducing carbohydrate synthesis (Badger et al.,
2000). Regarding temperature, a similar trend is observed due to
low temperature during the morning, with the normalized
photosynthesis rate due to temperature ranging from 0.65 to 0.63.
The optimal temperature for I. galbana T-iso has been reported to
be close to 36ºC (in this study), although other authors reported
values ranging from 25ºC to 30ºC (Renaud et al., 2002; Claquin et
al., 2008; Marchetti et al., 2012). Because the unfavourable
dissolved oxygen concentration, pH or temperature has a
multiplicative influence into the overall behaviour of the culture,
the final normalized photosynthesis rate can be calculated by
multiplying individual factors. Results shows as the overall
normalized photosynthesis rate range from 0.37 to 0.55, thus
60
indicating that biomass production achieved is approximately half
than the maximum achievable under optimal conditions.
To verify these phenomena the cultures light utilization efficiency
has been calculated on the basis of solar energy received and energy
stored into the biomass. Results show as the dilution rate has a
positive effect into the light use efficiency of the cultures by the
improvement of light availability however measured values are
lower than previously reported for other strains (Figure 22).
Fig. 22 - Variation of photosynthetic efficiency and quantum yield of Isochrysis galbana T-iso cultures with the dilution rate from experiments performed in 3.0 m3 industrial scale tubular photobioreactors.
Thus the photosynthetic efficiency ranged from 1.0 to 1.6 %,
whereas the quantum yield ranged from 0.2 to 0.3 g/E. Regarding
the photosynthetic efficiency, microalgae can conserve a maximum
of 10% of solar energy (photosynthetic efficiency) but outdoor
microalgal production systems rarely exceed 6% as yet (Carvalho et
al., 2006). The values reported here are lower than the maximal
values reported for outdoor cultures, 6.94% to 7.05% previously
0.0
0.1
0.2
0.3
0.4
0.5
0.0
0.2
0.4
0.6
0.8
1.0
1.2
1.4
1.6
1.8
0.10 0.20 0.30 0.40Qua
ntum
yield (g
/E)
Photosynthetic efficien
cy (%
)
Dilution rate (1/day)
Photosynthetic efficiency
Quatum yield
61
reported for Chlorella sp. cultures in thin-layer reactors (Doucha et
al., 2006; Doucha et al., 2009), or 3.6% reported for Scenedesmus
cultures in large tubular photobioreactors (Acién et al., 2012).
Regarding the quantum yield, measured values are lower than the
values of 0.5 g/E reported for Muriellopsis sp., 0.43 g/E reported
for P. subcapitata and 1.3 g/E reported for N. gaditana (Morales-
Amaral, et al., 2015; Sepulveda et al., 2015). For Isochrysis
galbana, a maximal value of 0.62 g/E was reported indoors
although this reduced to 0.1 g/E when calculated from outdoor
cultures (Molina et al., 1997).
62
5. CONCLUSIONS
Irradiance, temperature, pH and dissolved oxygen are relevant
variables determining the performance of Isochrysis galbana
cultures. In collaboration with the researchers at the “Universidad
de Almeria” a complete photosynthesis and respiration rate model
for Isochrysis galbana based on these variables was developed. The
model’s validity was verified outdoors using industrial-scale tubular
photobioreactors. The model allowed us to determine that
inadequate temperature and dissolved oxygen in the outdoor tubular
photobioreactors could reduce productivity by half that of the
maximal level according to light availability. The developed model
is a power tool for the design and management of Isochrysis
galbana-based outdoor processes, to take decisions about the
implementation of profitable control strategies. In addition, the
method used to construct this model is applicable to other strains
allowing us to optimize microalgae-based processes. Concerning
the production, the data confirm that Isochrysis galbana has a
growth rate lower than other strains. However, it is confirmed that
the photosynthesis rate can be doubled by optimizing the control of
the culture conditions in the reactors used.
Furthermore, this study showed that the productivity increases with
increasing dilution rate, so future studies considering higher rate
could be useful.
63
6. ACKNOWLEDGEMENTS
This research was supported by the Erasmus+ traineeship
programme of the Università di Bologna and the CO2ALGAEFIX
project (LIFE10 ENV/ES/000496) led by AlgaEnergy company. I
am most grateful to the Estación Experimental Las Palmerillas of
the Fundación Cajamar for collaborating in this research. This
research was supported by the Junta de Andalucía and the Plan